| The huge increase in the number of incarcerated people in the United States since 1980 has led to an unprecedented number of offenders returning to just a few disadvantaged inner-city neighborhoods. These offenders return with health problems, illiteracy, and little support from the justice system. Unfortunately, almost half return to prison or jail within three years, only to be released again with fewer job prospects and weaker ties to their community.;Using data on 38,411 felony convictions in Brooklyn over a 17½ year period, this dissertation examined the relationship between offender recidivism and neighborhood environments by examining the following research questions: (a) do recidivism hot spots exist?, (b) what are the social and physical characteristics of recidivism hot spots?, and (c) how important is the effect of local environment on recidivism as compared to traditional individual-level characteristics?;The first analysis confirmed the existence of six recidivism hot spots in Brooklyn. The second analysis focused on the characteristics of these recidivism hot spots, with the final regression model indicating that increased recidivism was significantly related to areas with fewer churches, fewer foreign born residents, fewer single parent households, fewer working mothers, and increased residential mobility. This model was significant at the .00001 level and accounted for 46.2% of the variability in tract-level recidivism.;The last analysis compared traditional individual-level variables to neighborhood-level variables using a multi-level model (offenders within census tracts). The final model contained six significant individual-level variables and two significant neighborhood-level variables, percent in poverty and density of all offenders, indicating that multiple neighborhood-level variables in the same model can have a significant affect on recidivism. Finally, the predictive power of a significant neighborhood-level variable (length of criminal career) was compared with that of a significant individual-level variable (poverty), and results indicated that a one year increase in the criminal career of an offender increases the of having a recidivism event by 66.96%, while a 10% increase in percent in poverty in a tract increases the odds of having a recidivism event by 4.4%. These results indicate that environment has a significant and measurable effect on recidivism. |
